Abstract:An effective source reconstruction method (SRM) to predict the noise coupling from an unknown electromagnetic interference (EMI) source inside a shielding enclosure is proposed in this work. On account of the difficulty and time-consuming in phase measurement of near-field, the magnitude-only near-field scanning is employed to obtain the near-field above EMI sources in free space, and the leakage near-field from the slots and/or holes of a shielding enclosure after the EMI source is inserted into the enclosure. The infinitesimal dipole is used as the equivalent source of the EMI source. The proposed method can generally be divided into three steps. First, the positions and initial moments of the equivalent dipoles of an unknown EMI source without any shielding enclosure are derived by using dynamic differential evolution (DDE) algorithm. Second, the obtained dipoles are thereafter placed into an arbitrarily shaped shielding enclosure, and the full-wave simulation is implemented to build the relationship between the dipoles and the measured leakage field from the slots/holes of the shielding enclosure. Third, the dipole moments are thereby optimized by using the DDE method again based on the magnitude-only leakage field. It is worth highlighting that only a relatively small region of near-field on the leakage plane is sufficient for obtaining an effective equivalent dipole model in the shielding enclosure. The results of numerical simulation and physical measurement have verified the effectiveness of the proposed SRM for modeling unknown EMI sources in a shielding enclosure.

A Deep Neural Network Modeling Methodology for Efficient EMC Assessment of Shielding Enclosures Using MECA-Generated RCS Training Data

An End-to-End Neural Network for Complex Electromagnetic Simulations

Shielding Effectiveness Characterization of Metallic Rectangular Cascaded Enclosures

EMI Radiation Prediction and Structure Optimization of Packages by Deep Learning

A Source Reconstruction Method for Unknown EMISources Inside a Shielding Enclosure Based Onmagnitude-Only Near-Field Scanning

Neural network model applied to electromagnetic shielding effectiveness of ultra-light Ni/Cu coated polyester fibrous materials

Machine Learning Methodology Review for Computational Electromagnetics

Hierarchical Attention-Based Machine Learning Model for Radiation Prediction of WB-BGA Package

Prediction Of Pcb Radiated Emission In Shielding Cavity Using Equivalent Dipole Modeling

An Equivalent Dipole Model Hybrid with Artificial Neural Network for Electromagnetic Interference Prediction

Reactive Near-Field to 3-Meter Far-Field Transformation Based on Deep Convolutional Neural Networks and Plane Wave Spectrum

AI Predictor: from Point Clouds to Scattered Far Fields for 3-D PEC Targets

Emi Prediction Of Packages By Deep Neural Network

A Hybrid Technique for Emi Prediction and Channel Modeling Inside an Enclosed Space

EMI Radiation Prediction of Packages Based on Convolution Neural Network

An Effective and Efficient Approach for Radiated Emission Prediction Based on Amplitude-Only Near-Field Measurements

Equivalent Electromagnetic Hybrid Dipole Based on Cascade-Forward Neural Network to Predict Near-Field Magnitude of Complex Environmental Radiation

Development of Artificial Neural Network for Field Prediction of Unknown EM Source

Visualization of Deep Convolutional Neural Networks to Investigate Porous Nanocomposites for Electromagnetic Interference Shielding

Fast Prediction for Electromagnetic Shielding Effectiveness of Ground-Via Distribution of SiP by Convolutional Neural Network

Efficient EMC Simulation of Enclosures with Apertures Residing in an Electrically Large Platform Using the MM-UTD Method